FACULTY PROFILE: Justin Williams
ecathletes must demonstrate skill in 10 events
(100 meter sprint, long jump, shotput, high jump, 400-meter run, 110
meter hurdle, discus throw, pole vault, javelin toss and 1,500 meter
run). They are considered among the best all-around athletes in the
world. Because rules require 30 minutes rest between events, decathletes
often spend their down time collaborating on ways to improve.
 |
Partially assembled
64-channel neural implant
(10K
JPG) |
A decathlete during his undergraduate career at South
Dakota State University, Justin
Williams also brings a well-rounded academic perspective to the
department. He holds bachelor’s degrees in mechanical engineering
and engineering physics, and received master’s and PhD degrees
in bio-engineering from Arizona State University. Before joining the
faculty as an assistant professor in summer 2003, he completed concurrent
postdoctoral fellowships in biomedical engineering at the University
of Michigan and neurosurgery at UW-Madison.
As a master’s student under Professors Daryl
Kipke and Andrew Schwartz at ASU, Williams combined his background in
mechanical engineering with his increasing interest in medical research.
“They had this idea that they could implant
some very small devices into the brain and record hundreds of neurons
and try to decode them,” says Williams.
The group hoped to use those decoded signals to drive
a robotic arm that someday would give people with amyotrophic lateral
sclerosis (ALS, or Lou Gehrig’s disease), high spinal-cord injuries
or brain-stem strokes a means of motion. Those patients’ brains
function, but because their motor output neurons are damaged, the information
that controls action is “locked” in. “We were in at
the ground floor,” says Williams. “It was something that
no one had ever tried and there were no wrong answers.”
As a mechanical engineer, Williams thought he would
contribute to designing the robotic arm. But by the time he finished
his PhD degree, he was developing the neural transmitters and refining
the surgical and biological steps for implanting them into the brain.
The technique worked: One of the monkeys, into which
Williams implanted neural transmitters, has learned to feed itself
by sending impulses via wires to an adjacent robotic arm. Now Williams’
goal is to apply this technology to humans and develop devices that
will function for the patient’s lifetime, will pose no significant
risk of infection or bleeding, and can be handled by a neurosurgeon.
“One of the problems we run into is the brain has a separate immune
system from the rest of the body and it acts much differently,”
he says. “It’s much more aggressive.”
The transmitters are so small—breathing on them
will scatter them like dust particles —that they elicit a response
from the body far different from that of a large device. “We thought
we could make things small enough that the body just doesn’t see
them,” says Williams. “And it turns out that if you make
something that’s as small as individual cells, then individual
cells attack them as if they were another foreign cell.”
To characterize the brain’s response to microscopic
implantable devices, Williams also develops novel noninvasive imaging
techniques. “A lot of histological methods are more suited for
doing large-scale reactions and they tend to destroy everything we’re
interested in,” he explains. He is working with teams around campus
to apply those findings and techniques to other situations, including
treating patients with Parkinson’s disease. UW-Madison doctors
treat 30 to 50 patients per year by implanting electrodes in their brains
to improve their motor control.
On any given day, Williams might collaborate with
neuroscientists, surgeons, physicians or other engineers to solve a
problem. To help prepare students to do the same, he is working with
BME and Medical School faculty to develop a PhD training program in
neuroengineering.
“We need to have people who speak a common
language,” he says.
